Stratosphere and Polar Vortex Watch

[Yarmy]

1 hour ago, feb1991blizzard said:

The difference between the EPS and GEFS remains stark.

Unfortunately though i get the feeling Micheal Ventrice's question is a rhetorical one.

At least next year the GEFS will be using the FV3 model, so we might not get led up the garden path so often.

[feb1991blizzard]

Just now, Yarmy said:

At least next year the GEFS will be using the FV3 model, so we might not get led up the garden path so often.

Yes - i think i may have to say hats off to @Recretos - He thinks the FV3 is decent at modelling the strat, now the EPS are more GFS op like and looks like the GEFS are on their own it looks like he is right

[Kirkcaldy Weather]

From Judah Cohen's latest blog 

"Therefore, a critical question for the winter weather and which forecast will favorably verify this winter – will there be an impactful PV disruption this winter?  The hemispheric pattern that set up in early to mid-November was highly favorable for disrupting the PV. With strong Ural/Scandinavian blocking and low heights near the Aleutians; and the PV has steadily but slowly weakened since the second week of November.  But since late November the Ural/Scandinavian blocking has been replaced by much lower pressures/heights (probably in part to the migration of the PV center to that region), a pattern less favorable for disrupting the PV.

The models in general have been less aggressive in disrupting the PV and probably a major mid-winter warming (reversal of the zonal mean zonal wind from westerly or positive to easterly or negative at 60°N and 10 hPa) is less likely than I previously thought.  All models predict a minor warming with ridging near Alaska and Northern Canada and the PV near and along the north slope of Eurasia.  This does resemble what is referred to as a Canadian warming.  This rarely results in a major warming but I do believe can be a precursor to a more significant warming later on.  In our paper on clustering of PV states (Kretschmer et al. 2018) the third cluster resembles a Canadian warming and the PV configuration for the first half of December with the PV displaced along the north slope of Eurasia and ridging warming near Alaska/Northern Canada.  I include in Figure ii the first figure from the paper showing the five different clusters and the NH temperature anomalies for each cluster.  The third cluster resembles both a Canadian warming and the near term forecast for the PV.  Also, the associated temperature pattern resembles the predicted NH temperature pattern with cold temperatures across Northern Asia and Europe but milder elsewhere across Eurasia.  The pattern across North America is reversed with relatively mild temperatures across Alaska and Northern Canada and seasonably cold across the US.

The GFS is predicting a relatively cold pattern across eastern North America for the next two weeks while the ECMWF is appreciably milder especially after day 10.  I hate to disagree with the superior ECMWF but it seems to me the GFS is more consistent with the temperature pattern associated with a Canadian warming.  Maybe like many model disagreements best to split the baby.

I believe that the extensive Siberian snow cover this October was a meaningful contributor to the PV weakening observed since mid-November and that is likely to continue.  However Siberian snow cover has not received the level of support from Barents-Kara sea ice to disrupt the PV as it has the past two winters.  Barents-Kara sea ice is close to normal and much higher than the two previous falls/early winters.  As an aside the strong warming bullseye over the Barents Kara Seas in the AER winter forecast may not verify with near normal sea ice in the region.

Snow cover by itself might not be able to disrupt the PV sufficiently to force a major warming this year.  Still based on the extensive snow cover, cold Siberia, generally low sea ice and warm Arctic I expect more perturbations to the PV in the coming months followed by periods of more severe winter weather.

Ironically though, low sea ice in the Barents-Kara Seas is favorable for disrupting the PV it does not seem to favor cold temperatures in the Eastern US.  Instead I believe that low sea ice in the Chukchi-Beaufort Seas and west of Greenland are more favorable for cold temperatures in the Eastern US.  Low sea ice in these regions support blocking near Alaska and Greenland respectively that often force troughing and cold temperatures in the Eastern US.  This winter, so far, the negative sea ice extent anomalies are greater in these regions relative to the Barents-Kara Seas

The plot of Wave Activity Flux (WAFz) or poleward heat transport shows strong negative anomalies this week followed by strong positive anomalies for the upcoming week (Figure 12). The predicted negative WAFz in mid-stratosphere possibly a sign of a reflective event.  In the weeks leading up to the major warmings in 2018 and 2019 there was no similar strong negative WAFz anomalies. Still it is confined to the mid-stratosphere which seems unusual to me.The stratospheric AO is currently slightly negative (Figure 1) reflective of a slightly perturbed PV.  However, in response to the positive WAFz predicted for next week, the stratospheric AO is predicted to turn more strongly negative (Figure 1) but likely not enough to qualify as a major warming (reversal of the zonal mean zonal wind from westerly or positive to easterly or negative at 60°N and 10 hPa).  

Despite the strong circulation around the PV center and relatively low heights, the PV is not circular in shape but rather elongated and displaced towards Eurasia, with the flow around the PV heavily skewed towards the Eurasian sector, signs of some disruption (Figure 13).  The counterclockwise flow around the PV center is bringing northerly flow to North Europe, supportive of the low geopotential heights and cold forecast for Scandinavia.

Currently there is warming and a ridge centered over Alaska and Northwest Canada in the stratosphere (Figure 13). Over time the new WAFz pulse is predicted to amplify the warming over the North Pacific side of the Arctic and reinforce the ridging centered over Alaska (Figure 13).  Also, the PV center is predicted to remain displaced towards northwest Eurasia over the next wo weeks.  The displacement of the PV center towards Scandinavia is likely contributing to a tropospheric reflection helping to deepen the troughing across Northern Europe next week (e.g., Figure 5).  The displacement of the stratospheric PV towards Eurasia is usually the first sign of a more significant PV disruption that is now being debated by most models." 

AO Blog Update | AER | Weather Risk Assessment

WWW.AER.COM

---

November 25, 2019 - Dr. Judah Cohen from Atmospheric and Environmental Research (AER) embarked on an experimental...

 

[mountain shadow]

After reading that, I don't think Cohen has anymore clue than us what the PV is going to do.

To many conflicting signals.

[Bricriu]

10 hours ago, mountain shadow said:

After reading that, I don't think Cohen has anymore clue than us what the PV is going to do.

To many conflicting signals.

He has been wrong in the past. I hope he is right about the Canadian Warming bit, though.

Edited December 3, 2019 by Bricriu

[bluearmy]

seems pretty safe to assess that the gefs are not worth using for guidance post day 10 in the upper strat 

 

[mb018538]

1 hour ago, bluearmy said:

seems pretty safe to assess that the gefs are not worth using for guidance post day 10 in the upper strat 

 

Or any GFS run to be honest...

Forecast:


Reality:


It's been over-cooking strat warmings and zonal wind-speed reductions all winter thus far. Best to not give them much serious thought until they get within a week if they are going to be so wrong.

[Summer Sun]

 

[Summer Sun]

 

[knocker]

 

[ArHu3]

Looks like the EC is picking up the beginning of another warming at the end of the run

Edited December 6, 2019 by ArHu3

[Interitus]

On 02/12/2019 at 19:25, feb1991blizzard said:

Yes - i think i may have to say hats off to @Recretos - He thinks the FV3 is decent at modelling the strat, now the EPS are more GFS op like and looks like the GEFS are on their own it looks like he is right

Here is a graph of the GFS and GEFS mean T384 forecast zonal wind compared to the actual analysis values, from November 15th onwards -

The GFS has forecast higher zonal wind speed, average +9.1 ms.

The ensemble mean has forecast lower zonal wind speed, average --4.2 ms. (Obviously a smoother plot as mean of 20 members). Interestingly these biases have been fairly consistent (i.e. above vs below) such that the average of the GFS and ensembles is closer than the two individually.

Must be remembered that this zonal wind is a pretty limited 1-dimensional quantity, compared to an areal analysis such as geopotential anomaly correlation - so forecast equal wind speeds which may appear accurate could be achieved from very different geopotential charts.

Case in point -

On 04/12/2019 at 09:34, mb018538 said:

Or any GFS run to be honest...

Forecast:


Reality:


It's been over-cooking strat warmings and zonal wind-speed reductions all winter thus far. Best to not give them much serious thought until they get within a week if they are going to be so wrong.

Here are the geopotential charts corresponding to the above temperature profiles -

The T384 forecast chart may have had the more noticeable warming posted above, zonal wind 28.6 ms, yet the second vortex of the 4th December analysis is more displaced and hence the vortex appears weaker from a zonal wind perspective at 18.1 ms. Not too much should be made of zonal wind or temperature charts in isolation.

Edited December 6, 2019 by Interitus

[Allseasons-Si]

@ArHu3

the gfs op/control are as well at the end but in diff locations.

lets see how it goes on later runs.

 

[feb1991blizzard]

3 minutes ago, Interitus said:

Here is a graph of the GFS and GEFS mean T384 forecast zonal wind compared to the actual analysis values, from November 15th onwards -

 

Seems like it all went badly wrong for the GEFS the last 5 days.

[Yarmy]

18 minutes ago, feb1991blizzard said:

Seems like it all went badly wrong for the GEFS the last 5 days.

From the 15th-20th November (subtract 384 hours from each date on the x-axis). It's interesting that the FV3 has a stronger positive bias than the GEFS has a negative bias at that range. Perhaps @Interitus could do the same chart for T240 lead times? 

 

[feb1991blizzard]

2 minutes ago, Yarmy said:

From the 15th-20th November (subtract 384 hours from each date on the x-axis). It's interesting that the FV3 has a stronger positive bias than the GEFS has a negative bias at that range. Perhaps @Interitus could do the same chart for T240 lead times? 

 

Yes - mind you i just hope we will be needing the charts again this winter!

[Interitus]

23 hours ago, Yarmy said:

From the 15th-20th November (subtract 384 hours from each date on the x-axis). It's interesting that the FV3 has a stronger positive bias than the GEFS has a negative bias at that range. Perhaps @Interitus could do the same chart for T240 lead times? 

 

Day 10 verification graph - analysis from Nov 9th 00z - Dec 7th 00z, and forecasts on to Dec 17th 00z

[Mike Poole]

5 hours ago, Interitus said:

Day 10 verification graph - analysis from Nov 9th 00z - Dec 7th 00z, and forecasts on to Dec 17th 00z

That is a really interesting analysis that you have done, there's been speculation about the relative performance of the FV3 and the GEFS on this thread for a few weeks, and even if there is no resolution about where we're going with the strat vortex, I think this proves some disconnect between what these models think,  we will be cautious from now on!  

(actually no we won't, the moment either the FV3 or GEFS mean shows support for a SSW, we will be it's cheerleaders...)

[Interitus]

1 hour ago, Mike Poole said:

That is a really interesting analysis that you have done, there's been speculation about the relative performance of the FV3 and the GEFS on this thread for a few weeks, and even if there is no resolution about where we're going with the strat vortex, I think this proves some disconnect between what these models think,  we will be cautious from now on!  

(actually no we won't, the moment either the FV3 or GEFS mean shows support for a SSW, we will be it's cheerleaders...)

It looks like the FV3 has the edge on purely simulating the strengthening Autumn vortex but then when the planetary waves really kick in from the middle of November. both tend to go a bit wayward. Don't know whether this is a wave propagation thing in the strat or due to the unpredictability of the waves from patterns in the trop.

[ArHu3]

EC is seeing a sharp drop in 1 hPa wind speeds, in previous years the lower levels always followed the 1hPa trend, so hopefully this trend continues 

[bluearmy]

3 hours ago, ArHu3 said:

EC is seeing a sharp drop in 1 hPa wind speeds, in previous years the lower levels always followed the 1hPa trend, so hopefully this trend continues 

It could hardly not drop though .....otherwise the upper atmosphere is in danger of escaping into deep space  !!!!!!

[Redbull165]

This is an absolute disaster, look at the zonal winds at the top of the stratosfere from 60-90N. The strong zonal-winds are progressing quickly into the lower parts of the troposfere. Which means an increase chance of the Atlantic (zonal-flow) for Europe dominating. I think we are about to lose the whole december, only a strong warming of the stratosfere with an increased wave1, and then wave 2 activity could change that. Otherwise the polar vortex will again regained its strenght and return to his roots at around Greenland. All we can do is hope, but for now it looks worse as in the start of 2018/19 winter. Maybe the second part of winter will bring something different.

And we can already see a crucial QBO (-EAST) flow around the equator from 1 hPa to 30 hPa. But usually the lag-time response from the QBO (west) to QBO (east) transitioning is at least about a month. So we might have to wait until january for the proper effect.

Edited December 8, 2019 by Redbull165

[Catacol]

1 hour ago, Redbull165 said:

This is an absolute disaster, look at the zonal winds at the top of the stratosfere from 60-90N. The strong zonal-winds are progressing quickly into the lower parts of the troposfere. Which means an increase chance of the Atlantic (zonal-flow) for Europe dominating. I think we are about to lose the whole december, only a strong warming of the stratosfere with an increased wave1, and then wave 2 activity could change that. Otherwise the polar vortex will again regained its strenght and return to his roots at around Greenland. All we can do is hope, but for now it looks worse as in the start of 2018/19 winter. Maybe the second part of winter will bring something different.

And we can already see a crucial QBO (-EAST) flow around the equator from 1 hPa to 30 hPa. But usually the lag-time response from the QBO (west) to QBO (east) transitioning is at least about a month. So we might have to wait until january for the proper effect.

To be fair - this is hardly a disaster. Next to no one looking at the winter in advance saw much chance of an icy December, and the vast majority of interpretations have suggested that we will need to wait for the second half of winter to produce the right combination of vortex disruption and pacific forcing to put the default westerly circulation under pressure. Look out for the impacts of MJO progression towards the end of the month in terms of blocking and possibly also vortex distress via renewed wave 2 forcing. The longer it remains pushed off its central axis through the middle part of this month the better. QBO interaction come end of January should also be favourable.

[swfc]

hi.bit of a novice here.i understand the mechanics of a ssw etc but just had a query ?Are they a seasonal thing ie "every year" or are they a symptom of certain atmospheric or synoptic circumstances? Tia

[Summer Sun]